Hxccptional Children VoL 71, No. 4. pp. 431-444. ©2005 Councilor Exceptional Childrtn.
Increasing Peer Interactions for Students With Severe Disabilities Via Paraprofessional Training N. CAUSTON-THEOHARIS KIMBER W. MALMGREN University ofWiicomin-Madison
ABSTRACT: As studcnts with severe disabilities are included in general education settings, the use of paraprofessionals has expanded to meet these students' needs. Unfortunately, paraprofessionals can have the inadvertent effect of intensifying the social isolation of students with disabilities. This study investigated the effectiveness of a training program aimed at teaching four paraprofessionals to facilitate interactions between students with severe disabilities and their peers. A multiple baseline, single-subject design across four paraprofessionallstudent pairs was utilized. Observational data were collected over the baseline and postintervention phases. Rates of paraprofessional facilitative behavior increased following the intervention. Additionally, rates of student interaction increased immediately and dramatically and were maintained through the maintenance probe.
T
he general education classroom has become the place where increasing numbers of students with disabilities are educated. Nationally, tbere are 5 1/2 million students with special needs, and slightly under half of these students in elementary schools are served in general education settings with their general education peers for more than 79% of the school day (U.S. Department of Education, 2002). Specifically, students with severe disabilities are included in general education settings with growing frequency, and increases arc predicted to continue as inclusionary practices become the norm (U.S. Department of Education).
Exceptional Children
A major impetus fot placing students with disabilities in general education classrooms is to allow them to reap the social and academic benefits afforded their peers without disabilities {Cullinan, Sabornie, & Crossland, 1992; Ferguson & Asch, 1989; Johnson & Johnson, 1991; Madden & Slavin, 1983; Wehman, 1990). Educational scholars have suggested that in an inclusive environment, being afforded the opportunity to learn from and care fot one another enriches the lives of students (Vandercook, Fleetham, Sinclair, & Tedie, 1998). The general education classroom is considered to be a fertile ground for the development of peet interactions and relationships. These peer interactions have been empirically linked to
431
The general education classroom is considered to be a fertile groundfor the development of peer interactions and relationships. increased achievement (Johnson, 1981; Yager, Johnson, & Johnson, 1985) and Increased selfesteem (Branthwaite, 1985; Kirova, 2001; Nave, 1990). However, for students with severe disabilities, these interactions and relationships may not occur naturally without appropriate support (Evans, Salisbury, Palombaro, Barryman, & Hollowood, 1992). The most common strategy that school districts use to support students with severe disabilities in inclusive classrooms is to allocate a paraprofessional to work with the individual student (Giangreco, Broer, & Edelman, 1999; Giangreco, Edelman, Broer, & Doyle, 2001; Werts, Wolery, Snyder, & Galdwell, 1996; Wolery, Werts, Caidweil, Snyder, & Liskowski, 1995). In some cases, the involvement of paraprofessionals may be the crucial support that allows a student with intensive academic or behavioral needs to be educated in a general education classroom or school rather than being placed in a more restrictive, segregated setting (Martella, MarchandMartella, Miller, Young, & Macfarlane, 1995). Although the assignment of a paraprofessional is intended to positively impact the stu-^ dent, several studies have shown that the presence of a paraprofessional can actually have detrimental effects on the peer interactions of a student with a disability (Giangreco, Edelman, Luiselli, & MacFarland, 1997; Marks, Schrader, & Levine, 1999; Shulka, Kennedy, & Cushing, 1999). Specifically, Giangreco et al. (1997) found that paraprofessional proximity had a profoundly negative impact on peer interactions, which affected relationships with classmates, Paraprofessionals were considered to be a physical barrier that caused many of the peers in the study to avoid the student with a disability. Giangreco et al. (1997) also reported that peers sometimes saw students and paraprofessionals as a "package deal." A second major ptoblem that students with disabilities who are supported by a paraprofessional face is separation from classmates (Gian-
greco et al., 1997). Paraprofessionals were routinely observed removing the students with disabilities from their peers or class grouping (e.g., moving the student to a back table to work or to another room without consultation with or tesistance from a teacher). Similarly, in a quahtative study by Malmgrcn and Causton-Theoharis (2003) of a student with emotional distutbance in an inclusive classroom, paraprofessional proximity was found to be the single most important classroom condition that negatively influenced peer interactions. A growing body of research documents that paraprofessionals are not well prepared to perform their specific job responsibilities (Brown, Farrington. Knight, Ross, & Ziegler, 1999; Giangreco et al., 2001; Wadsworth & Knight, 1996). It has been suggested that paraprofessionals who work with students with complex learning, cognitive, and behavioral issues are the least trained individuals in the school even though they are working with challenging students (Brown et al.). Many paraprofessionals receive no training before starting their employment in the schools (Passaro, Pickett, Latham, & HongBo, 1994), and many paraprofessionals report that they received the majority of their training by simply talking to and shadowing other paraprofessionals in the schools (Giangreco et al., 1997). Unfortunately, the support of an untrained paraprofessional can have negative consequences that actually undermine the original social and academic goals of inclusion. This study investigates the effectiveness of a training program aimed at teaching four paraprofessionals to facilitate interactions between students with severe disabilities and their peers. The research questions under investigation were as follows: Does training of paraprofessionals to facilitate interactions between students with and without disabilities increase the facilitative behaviors of the paraprofessionals? More important, does training of paraprofessionals increase the quantity of interactions that occur between students with disabilities and their peers? We predicted that the paraprofessional training would positively Influence both the behaviors of paraprofessionals and the interactions of students. Based on the fmdings from other researchers (e.g., Hundert & Hopkins, 1992), we further predicted that gains in facilitative behavior of paraprofes-
Summer 2005
sionals would level off at a lower point than gains In peer interaction, given that the goal of facilicative behavior is to encourage independent interactions thac blossom into even more interactions as students create meaningful relationships with their peers.
METHOD
SETTING
The study took place in two public elementary schools in a mid-size, midwestern school district. The school district was chosen because of its size and demonstrated commitment to inclusion. Students with disabilities throughout the district attended their neighborhood schools and were taught primarily in general education classrooms. The participating district employed 473 paraprofessionals during the 2002-2003 academic year. These paraprofessionals worked primarily with students with special needs. The specific elementary schools in which the study took place were solicited based on the enrollment of students with severe disabilities who were served primarily in general education classrooms wirh the support ofa paraprofessional. Two second-grade classrooms were utilized in "School A," and a kindergarten and a fourth-grade classroom were utilized in "School B." In each of the classrooms, a general educator was primarily responsible for the education of all students. The classes ranged in size from 15 to 22 students and the poverty rate, as determined by the percentage of students receiving free or reduced lunch, ranged from 10% to 25%.
For all four paraprofessionals, the data collection year was the 1st year each had worked with the specific target students. Two of the paraprofessionals had received no postsecondary education, and two had obtained bachelor's degrees. The paraprofessionals and the students in this study are referred to by pseudonyms. Paraprofessionals are referred to as Adele, Barb, Carla, and Don and the student participants in this study are referred to as Aivin, Barry, Charles, and Dustin. Student Participants. The student participants were four elementary students with severe disabilities who were supported by the paraprofessionals and who received the majority of their instruction (79% or more of the school day) in a general education setting. For the purposes of this study, the definition of a "severe disability" was taken from the Individuals with Disabilities Edu-
Paraprofessionah were considered to be a physical barrier that caused many of the peers in the study to avoid the student with a disability.
cation Act (IDEA) of 1997, meaning that the study population included students with significant learning or cognitive impairments who were also likely to have other accompanying physical or sensory impairments. The existence of concomitant impairments and their impact on communication, mobility, generalization, and/or major life activities was independently verified through perusal of each student participant's individualized education program (IEP). Two of the students in the study were identified by the school district as PARTICIPANTS having a primary disability label of autism; the Paraprofessional Participants. Study partici-other two were identified as having a primary dispants were comprised of four parapro- ability label of cerebral palsy. fcssional/student pairs. The paraprofessional All four of the students in this study were participants were recruited from among parapro- male and enrolled in elementary school. Two of fessionals who were primarily responsible for sup- the students were African American and two were porting a student with a severe disability in a Caucasian. Additional information about each of general education classroom. Three of the four the students' modes of communication and disparticipating paraprofessionals were female and all abilities is provided to give context to the issues were Caucasian, ranging in age from 35 to 53 that interfere with peer interaction. At tbe time of years. Their years of experience as a paraprofes- the study, Alvin, who was 7 years old and in the sional ranged from 3.5 to 7 years. second grade, had a vocabulary consisting of a
Exceptional Childrtn
few words (e.g., yes, no, swing, play), Alvin would, occasionally, get up and walk out of the room. This behavior was attributed to boredom by his teachers. Barry, an 8-year-old second grader, was able to talk; however, he spoke in one to three word utterances and his speech was sometimes difficuit to understand. When the study began, he knew over 50 words and concepts in sign language. When Barry was upset, he would sometimes yell or hit. Although he reportedly had never hit a peer, he frequently hit the paraprofessional who supported him. Charles, an 11-year-old fourth grader, spoke using one- or two-word utterances and a few signed words. When he was frusttated, he would breathe loudly or yell. Dtistin, a 6-year-old kindergarten student, communicated very well verbally; howevet, he had some problems with voice volume and eye contact. Additionally, Dustin used a wheelchair for mobility and a specialized supportive chair while doing schoolwork.
Many paraprofessionals receive no training before starting their employment in the schools.
INTERVENTION
Enhancing Perspective. Paraprofessionals were asked to complete a worksheet consisting of concentric circles (activity adapted from Forest, Pearpoint, & O'Brien, 1996). The paraprofessionals were asked to reflect on their own social relationships, indicating their own family and close friends in the innermost circle. Working outward through the circles, the paraprofessionals were then asked to write the names of good friends, the names of people they enjoyed doing things with occasionally, and last, the names of people who were paid to interact with them. The paraprofessionals were then directed to repeat this activity from the perspective of the target student with whom they worked. After they completed both sets of concentric circles, the paraprofessionals were prompted to compare the two resulting diagrams. The purpose of this activity was to enhance the perspective of the paraprofessional by providing a visual representation of the social relationships of the target student. In all four of the training sessions, the student circles generated by the pataprofessionals were virtually empty in the second and third tiers, but they were very full in the fourth, outermost circle (i.e., the tier indicating people who were paid to be with the student). By contrast, the circles of the paraprofessionals were much more balanced. Establishing the Importance of Peer Interac-
The intervention consisted of a 4-hour inservice tion. In the second activity, the trainer recorded training session held one-on-one with the partici- the responses while the paraprofessionals wete pating pataprofessionaJs. The curriculum used in asked "Why are social interactions and relationthe individual training sessions was entitled "Sup- ships important?" The paraprofessionals were porting Students with Disabilities in Inclusive then provided with information on this topic Schools" (Ghere, York-Batr, & Sommerness, from the training manual (e.g., "Friendships meet 2002). Unit 7 of the training program was used, as our human need to belong and feel cared about", it relates directly to the facilitation of interactions "Adults in schools can act as a bridge between stubetween students with and without disabilities. dents with and without disabilities", "Adults inThe first author conducted the individual training fluence whete, when, and how students spend sessions, meeting with each paraptofessional at his time together"). As this information was shared, or her place of employment for one session after the paraprofessionals were prompted to add to school. Training consisted of four activities with their own statements about the importance of sothe following ohjectives: (a) enhancing perspec- cial interactions and relationships for students. tive, (b) establishing the importance of peer interClarifying the Paraprofessional's Role in Facilaction, (c) clarifying the paraprofessionals role in itating Interactions. This activity involved underfacilitating interactions, and (d) increasing the scoring the paraprofessionals' responsibility to act paraprofessionals knowledge base of strategies for as a bridge between the target student and his or facilitating interactions. Each of these activities is her peers. The paraprofessionals were asked didescribed in more detail in the following. rectly to think of ways they could facilitate interSummn 2005
actions between target students and their peers. The trainer recorded their responses for later use.
rocal peer interactions between the participating students and other classmates. Increasing the Paraprofessional's Knowledge For the purpose of this study, faciUtative beBase. During this activity, strategies for facilitating haviors were defined as any purposeful behavior interaction were directly taught to the paraprofes- intended to cause the target student to interact sionals. These strategies included modeling ways with another student in the classroom. These beto interact, highlighting similarities between stu- haviors could include (a) increasing physical proxdents, identifying strengths of the target student, imity, (b) highlighting similarities, (c) teaching a directly teaching interaction skills, interpreting skill directly, and (d) modeling and interpreting behaviors, and actively partnering students. Ex- student behavior (Beckstead & Goetz, 1990). amples of each strategy (supplied in the training Peer interactions were defined as any two-way manual) were shared with the paraprofessionals. communication or any verbal or active nonverbal The paraprofessionals were then asked to add at behavior that causes another person to have a verleast one of their own ideas to each list of strategy bal or nonverbal response. This could include (a) examples and then talked through the possible questioning, (b) gesturing, (c) nodding, (d) carryapplication of each strategy to their own employ- ing out a direction, (e) physically or verbally rement situation. For a sample list of specific facilitative behaviors that were discussed in the training and later observed postintervention, see Figure 1. The first author concluded each training The paraprofessionals were asked directly session by verbally summarizing the list of facilitato think of ways they could facilitate intion strategies generated in the third and fourth activity. Subsequent to each training session, the teractions between target students and first author gave the typed list of compiled stratetheir peers. gies to the participating paraprofessional and the general and special education teachers with whom he or she collaborated. The intervention did not include any additional follow-up feedback or re- sisting an initiation, (f) significantly changing expression or making intentional eye contact, wards for the paraprofessional participants. and, (g) responding to or accepting physical support. If a second or further interaction was trigDATA COLLECTION PROCEDURES gered by the first, those ensuing interactions were The study was designed to evaluate the effective- counted as separate instances rather than being ness of the intervention on both the facilitative considered the continuation of a chain of behavbehaviors of the paraprofessionals and the rates of ior. The PIOI allowed the observer to record each interaction between the participating students facilitative behavior and peer interaction that ocand their peers during academic times. Parapro- curred during the set interval. Additionally, the fessional facilitative behaviors were documented instrument provided space for recording details through observation using the Peer Interaction about events or activities that might be pertinent and Paraprofessional Facilitative Behavior Obser- to the data (e.g., student resting head on desk). vation Instrument (PIOI). Student interaction Before data collection commenced, the first data were also collected via the PIOI. The PIOI author trained a second observer, a doctoral stuwas adapted from the Educational Assessment of dent in special education, in the use of the PIOI. Social Interaction (EASI) Engagement Scale Training continued until both observers simulta(Beckstead & Goetz, 1990), which was created to neously completed three consecutive 10-min obmeasure interactions between students with mul- servations with 100% agreement. tiple and severe disabilities and their peers. Individual observational probes were 10 min in DESIGN length. The PIOI was utilized to document the A multiple-baseline design across four paraprofesrate of the following occurrences; specific facilita- sional-student pairs was utilized. Observational tive behaviors of the paraprofessionals, and recip- data were collected during academic times for
Exceptional ChiUtren
FIGURE
1
Examples of Facilitative Behaviors Displayed by Paraprofessionals Postintervention Example Behaviors •
Increase target student's physical proximity to peers.
• Structure target student s "break time" to minimize removal from the classroom. • Redirect verbal queries about the target student directly to the student. • Fade assistance to allow more natural peer interaction opportunities. • Partner target student with peers during academic tasks. • Arrange for target student to use technology available in the classroom instead of in a separate setting. • Verbally highlight similarities between target student and peers. • Create communication cards focused on social exchanges for target student use. • Integrate target student's home experiences into classroom conversations. • Teach peers how to communicate with target student (e.g., selected ASL signs). • Directly teach peers and target students how to interact with one another. • Utilize interactive technology (e.g., computer with two input devices, tape player with two headsets). • Utilize rewards that are interactive in nature (e.g., lunch with a friend, puzzle time with a peer). • Give target student classroom responsibilities that encourage interaction (e.g., handing out papers). Note. ASL - American Sign Language each of the pairs over a 9-week period. Specifically, there was a 5-week period of ongoing data collection, a 4-week suspension of data collection, followed by two mamtenance probes. Baseline and Postintervention Phases. In order to establish the natural Frequency of the target behaviors of interest, baseline data on paraprofessional facilitative behaviors and peer interactions were collected fot a minimum of 3 observational days, or until stable baselines were established. Data were collected during consistent academic times in which the students in the classroom were expected to be learning new information or completing academic tasks and when interactions were appropriate. Observations were conducted three to eight times per week in each of the participating classrooms. Although the time of day that the observations took place varied because of the teachers' schedules, observations typically occurred sometime before lunch during the students' language arts block. After the intervention, data were continually collected until
the data trends were stable for all four pairs. Postintervention data collection procedures were identical to those utilized during baseline. Maintenance Probe. Four weeks after the last postintervention probe was completed, rwo maintenance probes were conducted in each classroom. Because the design necessitated that the intervention dates were staggered, this meant that the maintenance probe was conducted 8 weeks after the intervention occurred for the first pair, 7 weeks after the intervention occurred for the second pair, 6 weeks after the intervention occurred for the third pair, and 5 weeks after the intervention for the fourth pair. DATA DISPLAY AND
EVALUATION
Data were charted by plotting the rate of facilitation by the paraprofessional and the rate of student interaction for each observational period. The charted data were then evaluated by visual inspection (Kazdin, 1982). Data in this study were analyzed by examining changes in mean.
Summer 2005
level, and trend across the two phases (i.e., baseline and postintervention). Trend lines were created for each phase for each participant using the split-middle technique (White, 1972), and the percentage of nonovetlapping data between phases was calculated. INTEROBSERVER
AGREEMENT
facilirarive behaviors per min ranged from 0 to .1 facilitations across all four pairs. After the intervention, the rates of facilitative behaviors per min ranged from 0 to .7 across all four pairs. The change in range indicates that behavior after intervention became more variable. Overall, the paraprofessionals engaged in an average of two times more facilitative behaviors during the postintervention phase than during baseline. Although complete data categorizing the types of facilitative behaviors observed was not collected, anecdotal records indicated that partnering strategies most frequently yielded multiple interactions. During the baseline phase, no facilitative
Interobserver agreement checks were conducted during 18% (i.e., 24 out of 132) of the observational probes. These agreement checks were distributed equally across phases and participant pairs. During these checks, rwo observers independently collected data on each of the variables of interest and compared the data to determine to what extent the rwo data collectors agreed. A freOverall, the paraprofessionals engaged in quency ratio was calculated by comparing the total number of behaviors noted by the two ob- an average of two times more facilitative behaviors during the postintervention servers. The smaller number was divided by the larger and multiplied by 100 (Kazdin, 1982). phase than during baseline. Over the 24 simultaneous observations, the rate of interobserver agreement was 100% for both behaviors were observed for Adele. However, durvariables of interest. ing the postintervention phase, a rate of .14 (range 0 to .4) was recorded. Barb displayed an PROCEDURAL RELIABILITY average rate of .04 facilitative behaviors per min during baseline (range 0 to .1) and an average of In this study, all training sessions with paraprofes.29 (range 0 to .7) following the intervention. sionals were audio recorded. An independent Carla did not display any facilitative behaviors rater listened to the taped sessions to test fidelity during rhe baseline phase, whereas following the of treatment implementation by checking that all intervention, her average rate was .2 (range .1 to the steps in the training were completed during .4). Dons facilitative behavior average during each session. Review of the procedural reliability baseline was .004 (range 0 to .1), whereas his avdata revealed that 100% of the treatment compo- erage was .15 (range 0 to .3) following rhe internents were carried out during all four of the inde- vention. pendent training sessions. A maintenance probe administered 4 to 7 weeks after the completion of the postintervention phase reflected that each paraprofessional RESU LTS maintained his or her average level of facilitative behavior. Although the overall increase in facilitaPARAPROFESSIONAL FACILITATIVE tive behaviors was not dramatic, each of these BEHAVIOR maintenance probes demonstrated that the rate of Figure 2 presents the rates of facilitative behaviors facilitation remained at a level higher than the avper min for the four paraprofessional participants. erage baseline rate. Figure 2 graphically depicts Adele and Alvin are referred to as Pair #1, Barb the data generated from the maintenance probes. Trend lines depict the tendency for data to and Barry as Pair #2, Carla and Charles as Pair #3, and Don and Dustin as Pair #4. As depicted indicate systematic increases or decreases over in Figure 2, the rates of facilitative behaviors of time (Kazdin, 1982). In this study, trends in data the paraprofessionals increased slightly postinter- changed positively in slope for three of the four vention. During the baseline phase, rhe rates of paraprofessional participants immediately after
Exceptional Children
FIGURE
2
Rates of Paraprofessional Facilitative Behaviors and Student-Peer Interaction
Pair # I Adele & Alvin
Pair #2 Barb & Barry
Studenl Interactions
a
Facilitative Behaviors
CO
Pair #3 Ca«rla & Charles
13
CQ
u
Pair #4 Don & Dustin
\
A Observational Probes
Summer 2005
the intervention was employed. (See Figure 2 for postintervention phase trend lines; baseline phase trend lines for several participants either overlapped with or were just slightly above the x-axis and are therefore not depicted in the figure.) Examination of the trend lines indicated that each of the paraprofessionals increased their rate of facilitation following the intervention. For Carla, even though the trend line showed a deceleration, the overall rate of facilitation postintervention was much higher than would have been predicted from the baseline trend line. In addition, the maintenance probe for Carla showed a continued upward change.
During baseline, the rate of interaction for Alvin averaged .02 (range 0 to .1) per mm, whereas during postintervention the rate of interaction averaged .95 (range .2 to 2.9) per minBarry had an average interaction rate of .14 (range 0 to .4) during baseline, whereas following the intervention phase, his average interaction rate was 1.56 (range 0 to 1.9). Charles's interaction rate increased from an average of .06 (range 0 to .3) during baseline to 1.56 (range .7 to 2.0) during the postintervention phase. During baseline, the number of interactions for Dustin averaged .03 (range 0 to .3) per min; postintervention, Dustin's peer interactions averaged .53 (range .2 to .9) per min. Furthermore, mainThe percentage of nonoverlapping data be- tenance probes taken 4 weeks after the last obsertween phases was also calculated for each partici- vational probes demonstrated that rates of student pant. For Adele, a full 96.6% of the interaction remained relatively constant between postintervention data points were outside the the postintervention phase and the maintenance range of baseline data. For Barb and Caria those probes. percentages were 93.4 and 100 respectively. Postintervention data for Don showed a much lower percentage of nonoverlapping data, (i.e., 37.5%); On average, the students interacted 25 however, it is important to note that four of the times more frequently than during basefive overlapping postintervention data points line. comprised an overlap with the single non-zero baseline data point for this paraprofessional. The high percentages of nonoverlapping data (especially for Adele, Barb, and Carla), the immediate Inspection of the trend lines (see Figure 2) and obvious shifts in trend, and the overall in- for the student participants indicated that trends creased levels of performance (i.e., paraprofession- in the data changed positively in slope for three of als carrying out two times as many facilitative the four students. For Charles, the trend line behaviors postintervention), combined to create showed a slight deceleration; however, the change convincing evidence of change in facilitative be- in mean for this participant was still noteworthy. havior as a result of the intervention. The predicted visual continuation of trend lines from the baseline phases for all parricipating stuSTUDENT INTERACTION dents indicated that low levels of interaction would have been expected if no intervention had Figure 2 also depicts the rate of interactions per been employed. However, in all cases, the intermin between each of the target students and their vention caused the anticipated trend of the data peers. In all cases, the rate of interaction between to change drastically for the student participants. the target students and their peers increased when Even though Charles showed a slight deceleration rhe intervention was employed. During the basein his postintervention trend line, his trend line line phase, the rate of peer interactions for all four was still well above what would have been extarget students was very low, ranging from 0 to .4 pected without intervention, and the percentage interactions per min. During the postintervention of nonoverlapping data postintervention was 100. phase of the study, the rate ot interactions became much more variable, ranging from 0 to 2.9 per Percentage of nonoverlapping data points min. On average, the students interacted 25 times was also calculated for the other student particimore frequently than during baseline. pants. Like Charles, 100% of the postinterven-
Exceptional Children
tion data points were outside the range of Alvin's baseline range. For Barry and Dustin, the percentages of nonoverlapping data points were 88.9 and 75.0 respectively.
DISCUSSION
PARAPROEESSIONAL
FACILITATIVE
BEHAVIORS
Several prior investigators demonstrated that paraprofessionals can learn new skills and successfully implement new strategies after a relatively short training period (Hall, McClannahan, & Krantz, 1995; Shulka et al., 1999; Storey, Smith, & Strain, 1993; Wickham, 1993). The current study bolstered these findings and demonstrated that paraprofessionals can learn to facilitate interactions between students with and without disabilities after participating in a 4-hour inservice training. In this study, all four pataprofessionals increased their rates of facilitative behaviors after the intervention was employed. For example, Adele and Carla did not attempt to engage the target students in any interactive exchanges with other students during any observational period during the baseline phase. During the postintervention phase, these same two paraprofessionals increased their average rates of facilitation quite substantially. These two paraprofessionals went from displaying no faciiitative behaviors to engaging in one to two facilitations per 10 min probe on average. Another paraprofessional, Don, engaged in only one facilitative behavior during his very long baseline phase (i.e., 27 observational probes). After the intervention, he was observed employing more than one facilitative behavior on average during each observational probe. It is important to note that the remaining paraprofessional. Barb, demonstrated a slightly different pattern of facilitative behavior during the baseline phase than the other three paraprofessionals. Baseline data indicated that she actually did facilitate some peer interactions before the intervention was employed. She was observed facilitating a total of six interactions for Barry during the 17 observational probes that were conducted during baseline. However, like the other
paraprofessionals, she also increased her rate of facilitation from the baseline phase to postintervention—demonstrating that the training had a positive impact on a paraprofessional who demonstrated some level of proficiency, albeit limited, in facilitating interactions prior to the intervention. The change noted for Barb was the equivalent of going from an average of one facilitation every other observation, to almost three facilitations per 10 min observational period after the intervention. It is further important to note that Barb maintained the highest level of facilitation through the maintenance probes at the end of the study. After the intervention, paraprofessionals engaged in two times as many fecilitative behaviors as they did during baseline, resulting in natural and more frequent interactions between the target students and their peers. Additionally, paraprofessionals were then free to assist other students in the classroom while supervising the target student from a distance. Not only was this positive for other students in the classroom who received additional attention, it allowed for more natural peer interaction for the target student without an adult in the immediate vicinity. STUDENT INTERACTION
Several studies have documented positive benefits to students with severe disabilities placed in inclusive versus self-contained settings, including more interaction with peers without disabilities, IEP objectives related to social relationships, and higher levels of social support from peers (Fryxell
In many cases, one facilitative behavior resulted in numerous interactive exchanges between the target students and other students in the classroom. & Kennedy, 1995; Hunt, Farron-Davis, Beckstead, Curtis, & Goetz, 1994). Although the student participants in this study were most likely experiencing more interaction with their peers without disabilities than they would have experienced had they been served in segregated settings, they were still socially isolated during baseline. Although each target student had an IEP goal re-
Summer 2005
lated to social interaction, the paraprofessionals did very little to facilitate interactions duting the baseline phase. An important Finding in this study was that a relatively small change in paraprofessional behavior yielded a substantial increase in student interaction. In many cases, one facilitative behavior resulted in numerous interactive exchanges between the target student and other students in the classroom. For instance, Caria, one of the paraprofessionals, directed Charles to select a friend to read with him—an example of a partnering strategy presented in the paraprofessional training. The verbal directive was coded as one facilitative behavior. After this prompt, Charles selected a friend and sat down with her. Following Charles's selection, 14 interactive exchanges occurred over the next 5 min between Charles and his classmate. In another example. Barb {a paraprofcssional) and Barry were working together on a math assignment. When they finished, Barb noticed that another student in the classroom had also finished. Barb took that opportunity to utilize two facilitative strategies from the training. First, she highlighted the similarities between the two students by pointing out that they had both completed the same math worksheet. Second, she utilized a partnering strategy by suggesting that they check their answers together with a calculator. As they wotked together with the calculator, 19 peer interactions took place. Perhaps the most important result of this study was the dramatic and immediate change in the interaction levels observed between the target students and their peers. For all four student participants, rates of interaction prior to the intervention were extremely low. During the 10-min observational probes conducted during baseline, no target student engaged in more than four interactions. In fact, during baseline, tbe target students were not typically interacting with anyone at all. Of the 72 baseline probes, 61 (84%) reflected interaction rates of 0 or .1. This means that during 84% of the baseline probes, the target student was either not interacting at all, or engaged in the equivalent of one interaction over a 10-min period. This is drastically different than the interaction rates of students without disabilities in the same classrooms. These data show that the mere placement of students with disabilities
Exceptional Children
in general education classrooms does not ensure meaningful levels of interaction hetween students with and without disabilities. Our findings underscore Evans et al.'s (1992) assessment that someone within the school environment needs to facilitate and support the social inclusion of students within the classroom in order for that inclusion to be successful. As a result of the paraprofessional training, the interaction rates of the student participants began to approach the interaction rates of their peers. Peer interaction rates postintervention increased an impressive 25-fold. Additionally, though we did not attempt to record the valence of interactions observed, anecdotal notes recorded during observations suggest that only one interaction could have been considered negative. In that particular interaction, which occurred during baseline, a student took a ball from one of the target students, and the target student responded by shouting "No." The current study also supported Evans et al.'s (1992) fmdings in that when the paraprofessionals were trained to help students increase their interaction rates with peers, those rates did increase. Although the changes in paraprofessional facilitative behavior were not of the same magnitude as the changes in peer interaction, the changes in facilitative behavior and peer interaction occurred simultaneously and in conjunction with the intervention. The clear timing of the changes between the staggered phases makes us confident that facilitative behaviors learned in training made real changes in the classroom experiences of the participating students, with those students being much more meaningfully included in the classroom with their peers. SOCIAL VALIDITY
Whereas data concerning the social validity of the intervention were not formally collected, it is important to note that the parent of one student participant did independently contact the first author to let her know that she believed the intervention—and ensuing increases in social interaction and acceptance—had already made the classroom a more positive place for her son. She indicated that following the intervention, her son had twice been invited to play outside of school by classmates—his first such invitations of
the school year. Participating teachers and para- IMPLICATIONS FOR PRACTICE professionals also offered unsolicited testimonials about the strides they felt were made by the stu- The most common response to the increasing dent participants in terms of the overall quality of numbers of students with higii levels of need being served in inclusive settings is to hire paraprofestheir daily experiences in the classroom. sionals to support those students. The results of LIMITATIONS the current study show that this arrangement should be implemented with care. We recommend We acknowledge the existence of several limitathat when a paraprofessional is hired to support a tions to our findings. Although the data we colspecific student, he or she should be provided lected did show an increase in paraprofessional facilitative behavior after the intervention, that training, such as the intervention used in this increase was modest. Our hypothesis is that these study, to facilitate peer interactions. modest changes in adult behavior yielded notable Without proper training, paraprofessionals changes in levels of student interaction. However, can act in ways that unwittingly isolate and segreit is also possible that additional changes in paragate the students whom they support. This lack of professional behavior occurred that our data coladequate training has serious implications for the lection instrument was not sensitive enough to lives of students with disabilities, because interacdetect. An instrument focused on more subtle or qualitative changes in behavior or on a wider tion is essential to establish feelings of belonging range of behavior might have yielded different re- (Kunc, 2000; Maslow, 1970); self-esteem sults and should be considered in future studies of (Baumeister & Leary, 1995; Branthwaite, 1985; Nave, 1990); and improved academic success faciiitative behavior. (Johnson, 1981; Marr, 1997; Yager et al., 1985). The intervention package was multifaceted, The absence of interaction between the target stuaddressing both attitudes and perceptions about the importance of interactions, as well as specific dents and their peers during the baseline phase of skills to facilitate those interactions. Another limi- this study was consistent with other research doctation to our findings is that the observation in- umenting the negative effects of the use of parastrument measured global changes in facilitative professionals. Fortunately, these negative effects behavior only. Additional measures would be were diminished after the training was employed. This study demonstrated that a relatively needed to capture changes related to ail of the intervention components. We cannot assess at this short and low-cost paraprofessional training protime which parts of the training were most effec- gram could provide an immediate and potentially tive. Future research structured as a component long-lasting positive impact on the interaction analysis wouid allow the most effective combina- rates of students wirh severe disabilities in inclution of training package elements to be high- sive classrooms. In the future, this research should lighted. be expanded to other populations of students One other limitation to our findings is that with disabilities who also have difficulties wirh data were only collected during academic periods. peer interaction. In addition to our earlier recomWe know, however, that interaction during mendation that the utility of the specific compononacademic times is also critical to the academic nents of the training package be analyzed, we also and social success of students with severe disabili- recommend that the training program be impleties. Future research should focus on a wider mented on a wider scale to assess its broader funcrange of settings. tional utility and cost-effectiveness for professional development purposes. The results of this study and further research in this area can be Without proper training, paraprofessionused to facilitate the appropriate and supported als can act in ways that unwittingly isoinclusion of students with disabilities in general late and segregate the students whom they education classrooms so that their full academic support. and social potential can be reached.
Summer 2005
REFERENCES Baumeister, R. E, & Leary, M. R. (1995). The need to belong: Desire for interpersonal attachment as a fiandamental human morivation. Psychological Bulletin, 117. 497-529. Beckstead, S., & Goctz, L. (1990). EASI2 social interaction scale, V.6. San Francisco: California Research Institute. (ERIC Document Reproduction Service No. 365 049) Branthwaite, A. (1985). The development of social identity and self-concept. In A. Branthwaite & D. Rogers (Eds.), Children growing up (pp. 34-43). Philadelphia: Open University Press.
instructional assistant proximity on students with disabilities. Exceptional Children, 64, 7-18. Giangreco, M. E, Edelman, S. W., Broer, S. M., & Doyle, M. B. (2001). Paraprofessional support of students with disabilities: Literature from the past decade. Exceptional ChiL^ren, 68, 45-63. Hall, L. J., McClannahan, L. E., & Krantz, P. J. (1995). Promoting independence in integrated classrooms by teaching aides to use activity schedules and decreased prompts. Education and Training in Mental Retardation and Developmental Disabilities, 30, 208217.
Hundert, J., & Hopkins, B. (1992). Training supervisors in a collaborative team approach to promote peer Brown, L., Farrington, K., Knighr., T, Ross., C , & interaction of children with disabilities in integrated Ziegler, M. (1999). The need for fewer paraprofessionals and more teachers and i]\eri^\sts. Journal of the Asso- pteschools. Journal of Applied Behavior Analysis, 25, ciation for Persons With Severe Handicaps, 24, 249-252. 385-400. Cullinan, D., Sabornie, E. J., & Crossland, C. L. (1992). Social mainsrreaming of mildly handicapped students. The Elementary School journal, 92, 339-352.
Hunt, P, Farron-Davis, F, Beckstead, S., Curtis, D., & Goetz, L. (1994). Evaluating the effects of placement of students with severe disabilities in general education versus special classes. Journal of the Association Jor Persons With Severe Handicaps, 19, 200-214.
Evans, I. M., Salisbury, C. L., Palombaro, M. M., Barryman, J., & Hollowood, T. M. (1992). Acceptance of elementary-aged children with severe disabilities in an Johnson, D. W. (1981). Student-student interaction: inclusive school. Journal of the Association for Persons The neglected variable in education. Educational Researcher, 70(1), 5-10. With Severe Handicaps, 77,205-212. Ferguson, R, & Asch, A. (1989). Lessons from life: Personal and parental perspectives on school, childhood, and disability. In D. Biklin, A. Ford, & D. Ferguson (Eds.), Disability and society (pp. 108-140). Chicago: National Society for the Study of Education.
Johnson, D. W., & Johnson, R. T. (1991). Learning together and abne (3rd ed.). Englewood Cliffe, NJ: Allyn & Bacon. Kazdin, A. (1982). Single-case research designs. New York: Oxford University Press.
Forest, M., Pearpoint, J., & O'Brien, J. (1996). MAPs, Kirova, A. (2001, April). Social isolation, loneliness and Circles of Friends, and PATH: Powerful tools to help immigrant students' search jor belongingness: From helpbuitd caring communities. In S. Stainback & W. Stainlessness to hopefulness. Paper presented at the annual Inback (Eds.), Inclusion: A guide for educators (pp. 67-86). ternational Conference of the Association for Baltimore: Paul H. Brookes. Childhood Education, Toronto, Canada. Eryxell, D., & Kennedy, C. (1995). Placements along Kunc, N. (2000). Rediscovering the right to belong. In the continuum of services and its impact on students' R. A. Villa & J. S. Thousand, (Eds.), Restructuring for social relationships. Journal of the Association for Persons caring and effective education (pp. 77-92). Baltimore: With Severe Handicaps, 20, 259-269. Paul H. Brookes. Ghere, G., York-Barr, J., & Sommerness, J. (2002). N., & Slavin, R. (1983). Mainstreamlng stuSupporting students with disabilities in inclusive schools:Madden, A curriculum for job-embedded paraprofessional develop- dents with mild academic handicaps: Academic and social outcomes. Review of Educational Research, 53, ment. Minneapolis: University of Minnesota. 519-569. Giangreco, M. E, Broer, S. M., & Edelman, S. W. (1999). The tip of the iceberg: Determining whether Malmgren, K., & Causton-Theoharis, J. (2003). Boy in paraprofessional support is needed for students with the bubble: Effects of paraprofessional proximity and other disabilities in general education settings. Journal of the pedagogical decisions on the interactions of a student with behavioral disorders. Manuscript submitted for publicaAssociation jor Persons With Severe Handicaps, 24, 280tion. 290. Giangreco, M. E, Edelman, S., Luiseili, T. E., & MacFarland, S. Z. (1997). Helping or hovering? EfFects of
Exceptional Children
Marks, S. U., Schradet, G., &C Levine, M. (1999). Paraeducator experience in inclusive settings: Helping,
hovering, or holding their own. Exceptional Children, Wickham, D. R. (1993). An investigation of parapro65,315-328. fessional training to facilitate the inclusion of young Marr, M. B. (1997). Cooperative learning: A brief re- children with severe multiple disabilities and autism in community pteschools (mainstreaming). (Doctoral disview. Reading and Writing Quarterly, !3{l), 7-14. sertation. University of Kansas, 1993). Dissertation AbMartelia, R. C , Marchand-Martella, N. E., Miller, T stracts International, 55, Al 186. L, Young, K. R.. & Macfarlane, G. A., (1995). Teaching instructional aides and peer tutors to decrease prob- Wolery, M., Werts, M., Caldwell, N., Snyder, E., & lem behaviors in the classroom. TEACHING Liskowski, L. (1995). Experienced teachers' perceptions of resources and supports for inclusion. Education and Exceptional Children, 27(2), 53-56. Training in Mental Retardation and Developmental DisMaslow, A. H. (1970). Motivation and personality (2nd abilities, 30, 15-26. ed.). New York; HarperCollins. Yager, S., Johnson, D. W, & Johnson, R. T. (1985). Nave, B. (1990). Self-esteem: The key to student success. A Oral discussion, group to individual transfer, and series of solutions and strategies (Number 3). Clemson, SC: National Dropout Prevention Center. (ERIC Doc- achievement in cooperative ^royx^s. Journal of Educational Psychology, 77, 60-66. ument Reproduction Service No. 341919) Passaro, P, Pickett, A. L., Latham, G., & HongBo, W. (1994). The training and support needs of paraprofessionals in rural special education. Rural Special Education Quarterly, 13, 3-9. Shulka, S., Kennedy, C, H., & Cushing, L. S. (1999). Intermediate school students with severe disabilities: Supporting their social participation in general education classrooms. Journal of Positive Behavior Interventions, i, 130-140.
ABOUT THE JULIE
AUTHORS
N. CAUSTON-THEOHARIS (CEC
Chapter #14), Lecturer; and K I M B E R W .
MALMGREN (CEC Chapter #832), Assistant Professor, Department of Rehabilitation Psychology and Special Education, University of Wisconsin-Madison.
Storey, K., Smith, D. J., & Strain, P. S. (1993). Use of classroom assistants and peer-mediated intervention to increase integration in preschool settings. Exceptionality, Julie N. Causton-Theoharis is now at Syracuse 4, 1-16.
University.
U.S. Department of Education. (2002). Twenty-fourth annual report to Congress on the implementation of theAddress correspondence to Julie N. CaustonIndividuals with Disabilities Education Act. Washington, Theoharis, Syracuse University, Department of DC: U.S. Government Printing Office. Vandercook, T, Fleetham, D., Sinclair, S., & Tetlie, R. (1998). Cath, Jess, Jules and Ames . . . A story of friendship. IMPACT, 2, 18-19. Wadsworth, D. E., & Knight, D. (1996). Paraprofessionals: The bridge to successful full inclusion. Intervention in School and Clinic, 31, 166-171.
Teaching and Learning, 150 Huntington Hall, Syracuse, NY 13244-2340 (e-mail:
[email protected] [until July 2005] or to Kimber Malmgren at kmalmgren@education. wisc.edu) Manuscript received October 2003; accepted
Wehman, P (1990). School to work: Elements of suc- April 2004. cessful programs. TEACHING Exceptional Children, 23(1), 40-43. Werts, M. G., Wolery, M., Snyder, E. D., &c Caldwell, N. K. (1996). Teachers' perceptions of the supports critical to the success of inclusion programs. Journal of the Association jor Persons With Severe Handicaps, 21, 921. White, O. R. (1972). A manual for the calculation and use of the median slope—A technique of progress estimation and prediction in the single case. Eugene: University of Oregon, Regional Resource Center for Handicapped Children.
Summer 2005
